Skip to content

MCP & tool use

How models connect to the outside world: the MCP ecosystem, function calling and tool integrations.

824 picksRelated topicsAgentsAI codingOpen source

Latest picks

721–740 of 824

Feb 14Saturday

Ruan YiFeng's Weblog

Using ByteDance's Seed 2.0 and TRAE with Skills for app building and deployment

Ruanyifeng used ByteDance's Seed 2.0 Code and TRAE to generate one ASCII-to-Excalidraw web app and preview it at localhost:8080. The post says Seed 2.0 includes Pro, Lite, Mini, and Code models, and shows Skills as YAML-headed Markdown files, including Anthropic's frontend-design and Vercel deploy examples.

Why it matters: Ruan Yifeng used ByteDance's Seed 2.0 Code model and the TRAE tool to build a web app that turns ASCII art into hand-drawn style images, and explains how Skill files reuse coding agent workflows.

MIT Technology Review · AI

ALS stole this musician’s voice. AI let him sing again.

Patrick Darling, 32, returned to the stage on February 11 in London after two years without singing, using an AI voice clone rebuilt from old recordings. The post says speech cloning typically needs about 10 minutes of clean audio; his singing clone was built from noisy phone clips and kitchen recordings, then refined with Eleven Music over about six weeks. The practical signal is access, not sentiment: ElevenLabs offers the tools free to people who lost their voices to ALS and similar conditions, but the post does not disclose model details.

Why it matters: A 32-year-old musician with ALS sang with his band again using an AI voice clone built from noisy phone video and kitchen recordings.

Feb 12Thursday

Lex Fridman (YouTube RSS)

OpenClaw: The Viral AI Agent Behind the Hype - Peter Steinberger | Lex Fridman Podcast #491

Lex Fridman’s episode #491 interviews Peter Steinberger about the open-source AI agent OpenClaw; the transcript says it reached 175k-180k GitHub stars. The post says it can connect to Telegram, WhatsApp, Signal, and iMessage, and use models such as Claude Opus 4.6 and GPT 5.3 Codex; it does not fully disclose the architecture, evals, or security boundaries. The real point is system-level access and self-modifying behavior: this is not chat, but an agent that can take actions.

Why it matters: OpenClaw creator Peter Steinberger tells Lex Fridman how the open-source AI agent went viral, covering its ability to pass CAPTCHAs, edit code and run inside everyday chat apps.

MIT Technology Review · AI

Is a secure AI assistant possible?

OpenClaw was uploaded to GitHub in November 2025 and went viral in late January, extending LLMs into email, browsing, and local files with larger security risks. The post names prompt injection as the central threat, says there are likely “hundreds of thousands” of OpenClaw agents online, and notes a public warning from the Chinese government. The key point: the article says there is no silver-bullet defense yet, and the truncated body does not disclose the full mitigation details.

Why it matters: Once OpenClaw connects a model to email, a browser and local files, prompt injection lets an attacker hijack the agent, and the piece asks where such an assistant's security boundary sits.

Feb 10Tuesday

36Kr (direct RSS)

OpenAI to integrate ChatGPT into the U.S. Department of Defense's generative AI platform

The U.S. Department of Defense will work with OpenAI to integrate ChatGPT into GenAI.mil for about 3 million personnel. The RSS snippet discloses the integration, platform name, and user count, but not the model version, access controls, scope, or launch date.

Why it matters: The US Department of Defense will plug ChatGPT into its internal generative AI platform GenAI.mil, reaching about 3 million personnel.

36Kr (direct RSS)

Embodied AI company Noematrix raises several hundred million yuan in Series A, with overseas funds joining

Noematrix closed a Series A worth several hundred million yuan, led by C Capital, with Sea Limited and Puhua Capital participating, and Prosperity7 Ventures increasing its stake. Founded in Nov. 2023, the company says its Noematrix Brain has been deployed on wheeled single-arm, wheeled dual-arm, and humanoid dual-arm robots in retail pharmacies and hotel laundries; the post does not disclose valuation or revenue. The sharper signal is its claimed hundreds of thousands of hours of real-robot data and its data-model-scenario loop.

Why it matters: A robotics brain maker raised several hundred million yuan in Series A, led by C Capital, with new backers including Sea Limited and existing investors following on.

Feb 9Monday

36Kr (direct RSS)

Voice Ask is live: why is Xiaohongshu pushing search-by-question?

Xiaohongshu fully launched Voice Ask on Jan. 27, letting users long-press to speak on the search page and get structured answers distilled from in-app user experience posts. The post says it can handle 3-minute spoken queries, foreign languages, and dialects, but does not disclose the model, ASR stack, latency, or accuracy. The real shift is from 3-4 character keyword search to longer spoken questions, widening search intent capture and scenario coverage.

Why it matters: Xiaohongshu rolled out voice Q&A to all users, and answers to long-pressed voice questions on the search page are limited to real user experience on the platform rather than general encyclopedic results.

36Kr (direct RSS)

Qwen’s 10 Million Milk Teas: How Alibaba’s Massive AI Freebie Campaign Unfolded

Alibaba’s Qwen drove over 10 million orders via a Feb. 6 free-order campaign, but the app slowed and crashed from 10 a.m. to noon as load exceeded capacity; orders had already passed 2 million before noon. 36Kr says initial server capacity was only about one-third of the expected peak, and the subsidy pool was framed as 3 billion yuan; the real signal is not a model leap but a paid test of AI commerce entry and consumer acquisition.

Why it matters: A live look at what happens when an AI app's traffic hits three times the capacity it planned for.

Feb 7Saturday

MIT Technology Review · AI

Moltbook was peak AI theater

Moltbook went viral within hours, and the platform says it now has 1.7 million agent accounts, 250,000 posts, and 8.5 million comments, but the article argues the activity is mostly human-scripted mimicry. It says OpenClaw can connect Claude, GPT-5, or Gemini to tools like email and browsers; cited operators say the agents lack shared goals, shared memory, and self-directed autonomy, and some viral posts were written by humans posing as bots. The key takeaway is risk: agents tied to private data such as passwords or bank details were active on a site filled with spam and potentially malicious instructions.

Why it matters: Within hours of launch, Moltbook gathered 1.7 million bot accounts, 250,000 posts and 8.5 million comments, with the bots run automatically by OpenClaw wiring large models into email and browsers.

Feb 6Friday

TechCrunch · AI

OpenAI launches new agentic coding model minutes after Anthropic releases its own

OpenAI launched an agentic coding model minutes after Anthropic released a similar one, and the model is meant to accelerate Codex, which OpenAI launched earlier this week. The RSS snippet gives only the timing and purpose; the post does not disclose the model name, benchmarks, pricing, context length, or availability. The signal is direct competition in agentic coding, not a substantiated performance claim.

Why it matters: OpenAI shipped GPT-5.3 Codex minutes after Anthropic released a comparable model, as both firms bet on coding models that finish development tasks on their own.

Feb 4Wednesday

TheValley101 (硅谷101)

E224 | Why Clawdbot became the first breakout product of 2026 amid the Mac mini rush | Moltbot | MoltBook | OpenClaw

The podcast says Clawdbot passed 100k GitHub stars within days and reached 146k on Feb. 2, while being renamed to Moltbot and then OpenClaw within a week. It attributes the traction to a stack of Claude, long-term memory, IM-based messaging, and proactive heartbeat workflows; the title mentions a Mac mini rush, but the post does not disclose sales figures. The real signal is the interaction layer rather than a new model release: this is industry commentary and user anecdotes, not an official spec sheet.

Why it matters: This podcast breaks down Clawdbot's surge from user, software and hardware angles, then tracks the reversals that followed: the rename, the lawsuit, exposed security holes and servers going down.

Feb 3Tuesday

Computing Life · Yage

Beyond Tutorial Thinking: Why AI Education Should Add Engineering Infrastructure, Not Just Content

The team says it ran 4 courses over 2 years for 2,500+ learners, yet only a minority shipped usable products; drop-off centered on setup, experimentation, deployment, and context handling friction. The post says AI Builder Space gives students a no-card unified API, one-click deployment to <name>.ai-builders.space free for 1 year, and MCP access for Cursor and Claude Code via one command. The point is productized teaching infra, not more tutorials; retention, conversion, and cost are not disclosed.

Why it matters: A team running four courses for 2,500+ students found most got stuck on card binding, setup and deployment, so it built a teaching platform with a no-card unified API.

MIT Technology Review · AI

What We’ve Been Getting Wrong About AI’s Truth Crisis

MIT Technology Review says the US Department of Homeland Security has confirmed using Google and Adobe AI video generators for public-facing content, reported last Thursday. The post cites two failure points: Adobe auto-labels only fully AI-made content, mixed edits are opt-in, and X can remove or hide labels. The key issue is influence after exposure: a new Communications Psychology paper found participants still used a fake confession deepfake to judge guilt even after being told it was fake.

Why it matters: MIT Technology Review uses a DHS promotional video made with AI video tools and experiments showing people still act on videos they know are fake to reframe where the AI forgery crisis actually bites.

Jan 30Friday

Ruan YiFeng's Weblog

Technology Enthusiast Weekly #383: What Level of AI Programming Are You?

Steve Yegge frames AI coding into 8 levels and says he is at level 8, where an orchestrator manages parallel AI coding sessions. The post lays out a path from IDE copilots to YOLO acceptance, 3-5 windows, 10+ windows, then orchestration; it also says his AI-built tool Gas Town has 225,000 lines of Go code, which he has never read, and had 6,000 stars as of last week. The real signal is black-box programming as a workflow choice, with cost and failure risk stated plainly.

Why it matters: A concrete ladder for AI coding autonomy, with the top rung claimed at 225,000 lines of Go written and never read.

Bloomberg Technology

Perplexity Inks Microsoft AI Cloud Deal Amid Dispute With Amazon

Perplexity signed a $750 million Azure cloud deal with Microsoft while facing a legal dispute with its longtime cloud partner Amazon. The RSS snippet discloses the deal size, cloud provider, and dispute context, but not the contract term, compute scale, or lawsuit details. The key signal is a cloud supply rebalance that can affect training and inference costs.

Why it matters: Perplexity signed a $750 million Microsoft Azure cloud contract while a long-time AWS user, against the backdrop of a legal dispute with AWS, shifting the cloud supply picture.

Bloomberg Technology

Amazon in Talks to Invest Up to $50 Billion in OpenAI and Expand Ties

Amazon is in talks to invest up to $50 billion in OpenAI and expand their existing relationship. The RSS snippet says the tie-up includes Amazon selling compute to OpenAI; the post does not disclose deal structure, timing, or whether talks will close. The key signal is compute linkage, not just capital.

Why it matters: Amazon is negotiating an investment of up to $50 billion in OpenAI while also selling it compute, tying the money to cloud services.

MIT Technology Review · AI

DHS is using Google and Adobe AI to make videos

A DHS document says the agency uses Google Veo 3, Google Flow, and Adobe Firefly for public-facing content, with an estimated 100 to 1,000 licenses. It also says DHS uses Microsoft Copilot Chat for drafting and summarization and Poolside for coding; the post does not disclose which specific videos used which tool. The key point for practitioners is that commercial video generators are now inside a federal public-communications workflow, while watermark retention and attribution remain unverifiable across platforms.

Why it matters: A published DHS list of AI tools shows its external video production already runs on Google Veo 3, Google Flow and Adobe Firefly, under licenses numbering between 100 and 1,000.

Jan 28Wednesday

Mistral AI

Mistral releases terminal coding agent Mistral Vibe 2.0

Mistral released Mistral Vibe 2.0, a terminal coding agent powered by the Devstral 2 model family. It adds custom subagents, multi-option clarification, slash-command skills, a unified agent mode and automatic updates.

Why it matters: Mistral's terminal coding agent adds custom subagents, multi-option clarification and slash-command skills, plus token-based API pricing for the Devstral 2 line.

Jan 27Tuesday

MIT Technology Review · AI

Inside OpenAI’s big play for science

OpenAI launched its OpenAI for Science team in October 2025 to test how GPT-5-class models can support scientists. Kevin Weil said GPT-5.2 scored 92% on GPQA versus GPT-4’s 39%; the piece also notes OpenAI deleted posts that overstated old-paper retrieval as solving unsolved math problems.

Why it matters: OpenAI's new team, led by Kevin Weil, is testing whether GPT-5 can help scientists generate ideas, pick directions and surface obscure old papers, defining the model's role in research.

Jan 23Friday

MIT Technology Review · AI

“Dr. Google” had its issues. Can ChatGPT Health do better?

OpenAI launched ChatGPT Health this month, and says 230 million people ask ChatGPT health questions each week. The post says it is not a new model but a wrapper with health guidance and tools, including optional access to medical records and fitness data. The real issue is evaluation: cited studies put GPT-4o at about 85% accuracy on realistic prompts, but only about half of no-choice licensing answers were rated fully correct.

Why it matters: OpenAI launched ChatGPT Health, 230 million people a week already ask ChatGPT health questions, and one study found experts judged only about half of GPT-4o's medical licensing answers fully correct.